Ross Wightman 9bcd65181b Add exponential moving average for model weights + few other additions and cleanup ...

* ModelEma class added to track an EMA set of weights for the model being trained
* EMA handling added to train, validation and clean_checkpoint scripts
* Add multi checkpoint or multi-model validation support to validate.py
* Add syncbn option (APEX) to train script for experimentation
* Cleanup interface of CheckpointSaver while adding ema functionality

2019-06-07 15:39:36 -07:00

..

__init__.py

Uniform pretrained model handling.

2019-04-11 21:32:16 -07:00

adaptive_avgmax_pool.py

Fix pooling in mnasnet, more sensible default for AMP opt level

2019-04-17 18:06:37 -07:00

conv2d_same.py

Add support for tflite mnasnet pretrained weights and included spnasnet pretrained weights of my own.

2019-04-28 17:45:07 -07:00

densenet.py

Cleanup model_factory imports, consistent __all__ for models, fixed inception_v4 weight url

2019-05-28 21:41:10 -07:00

dpn.py

Cleanup model_factory imports, consistent __all__ for models, fixed inception_v4 weight url

2019-05-28 21:41:10 -07:00

gen_efficientnet.py

Ported Tensorflow pretrained EfficientNet weights and some model cleanup

2019-05-30 17:55:35 -07:00

gluon_resnet.py

Cleanup model_factory imports, consistent __all__ for models, fixed inception_v4 weight url

2019-05-28 21:41:10 -07:00

helpers.py

Add exponential moving average for model weights + few other additions and cleanup

2019-06-07 15:39:36 -07:00

inception_resnet_v2.py

Cleanup model_factory imports, consistent __all__ for models, fixed inception_v4 weight url

2019-05-28 21:41:10 -07:00

inception_v3.py

Cleanup model_factory imports, consistent __all__ for models, fixed inception_v4 weight url

2019-05-28 21:41:10 -07:00

inception_v4.py

Cleanup model_factory imports, consistent __all__ for models, fixed inception_v4 weight url

2019-05-28 21:41:10 -07:00

median_pool.py

Initial commit, puting some ol pieces together

2019-02-01 22:07:34 -08:00

model_factory.py

Ported Tensorflow pretrained EfficientNet weights and some model cleanup

2019-05-30 17:55:35 -07:00

pnasnet.py

Cleanup model_factory imports, consistent __all__ for models, fixed inception_v4 weight url

2019-05-28 21:41:10 -07:00

resnet.py

Replace ResNet-34 default weights with a great result from my experiments.

2019-05-31 10:42:46 -07:00

senet.py

Missed a few models in the model/_all_ list for senet

2019-05-31 10:51:18 -07:00

test_time_pool.py

Add per model crop pct, interpolation defaults, tie it all together

2019-04-12 22:55:24 -07:00

xception.py

Cleanup model_factory imports, consistent __all__ for models, fixed inception_v4 weight url

2019-05-28 21:41:10 -07:00